Mechanochemical and solution synthesis, X-ray structure and IR and 31P solid state NMR spectroscopic studies of copper(I) thiocyanate adducts with bulky monodentate tertiary phosphine ligands.

A number of adducts of copper(I) thiocyanate with bulky tertiary phosphine ligands, and some nitrogen-base solvates, were synthesized and structurally and spectroscopically characterised. CuSCN:PCy3 (1:2), as crystallized from pyridine, is shown by a single crystal X-ray study to be a one-dimensional polymer ...(Cy3P)2CuSCN(Cy3P)2CuSCN... (1) with the four-coordinate copper atoms linked end-on by S-SCN-N bridging thiocyanate groups. A second form (2), obtained from acetonitrile, was also identified and shown by IR and 31P CPMAS NMR spectroscopy to be mononuclear, with the magnitude of the dν(Cu) parameter measured from the 31P CPMAS and the ν(CN) value from the IR clearly establishing this compound as three-coordinate [(Cy3P)2CuNCS]. Two further CuSCN/PCy3 compounds CuSCN:PCy3 (1:1) (3), and CuSCN:PCy3:py (1:1:1) (4) were also characterized spectroscopically, with the dν(Cu) parameters indicating three- and four-coordinate copper sites, respectively. Attempts to obtain a 1:2 adduct with tri-t-butylphosphine have yielded, from pyridine, the 1:1 adduct as a dimer [(Bu(t)3P)((SCN)(NCS))Cu(PBu(t)3)] (5), while similar attempts with tri-o-tolylphosphine (from acetonitrile and pyridine (= L)) resulted in solvated 1:1:1 CuSCN:P(o-tol)3:L forms as dimeric [{(o-tol)3P}LCu((SCN)(NCS))CuL{P(o-tol)3}] (6 and 8). The solvent-free 1:1 CuSCN:P(o-tol)3 adduct (7), obtained by desolvation of 6, was characterized spectroscopically and dν(Cu) measurements from the 31P CPMAS NMR data are consistent with the decrease in coordination number of the copper atom from four (for 6) (P,N(MeCN)Cu,S,N) to three (for 7) (PCuS,N) upon loss of the acetonitrile of solvation. These results are compared with those previously reported for mononuclear and binuclear PPh3 adducts which demonstrate a clear tendency for the copper centre to remain four-coordinate. The IR spectroscopic measurements on these compounds show that bands in the far-IR spectra provide a much more definitive criterion for distinguishing between bridging and terminal bonding than does an often-used empirical rule based on ν(CN) in the mid-IR, which leads to the wrong conclusion in some cases.